To address the clinical concerns of immunogenicity and suboptimal pharmacokinetics, cancer therapy with monoclonal antibodies has evolved from murine to chimeric, humanized, and fully human constructs. Parallel to these improvements have been continuing efforts to develop more effective forms of antibodies, which to date include different antibody isotypes, single-chain antibody fragments with monomeric or multimeric binding moieties, specific mutations in the Fc region to modulate effector function or circulating half-life, and bispecific antibodies of numerous designs that vary in valency, structure, and constituents (Chames et al., Br J Pharmacol 2009, 157:220-233).
Because signaling pathway redundancies can result in lack of response to a single antibody, diverse strategies to use combination therapy with antibodies that bind to different epitopes or different antigens on the same target cell have been proposed. Combinations such as anti-CD20 and anti-CD22 (Stein et al., Clin Cancer Res 2004, 10:2868-2878), anti-CD20 and anti-HLA-DR (Tobin et al, Leuk Lymphoma 2007, 48:944-956), anti-CD20 and anti-TRAIL-R1 (Maddipatla et al., Clin Cancer Res 2007, 13:4556-4564), anti-IGF-1R and anti-EGFR (Goetsche et al., Int J Cancer 2005, 113:316-328), anti-IGF-1R and anti-VEGF (Shang et al., Mol Cancer Ther 2008, 7:2599-2608), or trastuzumab and pertuzumab that target different regions of human EGFR2 (Nahta et al., Cancer Res 2004, 64:2343-2346) have been evaluated preclinically, showing enhanced or synergistic antitumor activity in vitro and in vivo.
The first clinical evidence of an apparent advantage of combining two antibodies against different cancer cell antigens involved the administration of rituximab (chimeric anti-CD20) and epratuzumab (humanized anti-CD22 antibody) in patients with non-Hodgkin lymphoma (NHL). The combination was found to enhance anti-lymphoma efficacy without a commensurate increase in toxicity, based on 3 independent clinical trials (Leonard et al., J Clin Oncol 2005, 23:5044-5051).
Given the number of antibodies approved for cancer therapy, the number of such potential combinations is not large. However, where such combinations show improved efficacy, there is concern over the combined cost of individually expensive antibody therapies, in addition to the inconvenience and time of conducting separate infusions. As an alternative, attempts to develop bispecific antibodies that can simultaneously bind two target antigens have resulted in a multitude of approaches (Chames & Baty, Curr Opin Drug Discov Devel 2009, 12:276-283).
Earlier methods used for the production of bispecific antibodies made use of chemical cross-linking of IgG or Fab′ (Perez et al., Nature 1985, 316:354-356; Glennie et al., J Immunol 1987, 139:2367-2375) or quadromas obtained by fusing two hybridomas together (Staerz & Bevan, Proc Natl Acad Sci USA 1986, 83:1453-1457). Subsequent strategies focused on generating recombinant bispecific antibodies composed of tandem scFvs or diabodies (Kriangkum et al., Biomol Eng 2001, 18:31-40). One format of such Fc-lacking constructs, referred to as BiTe, is currently being tested clinically (Baeuerle & Reinhardt, Cancer Res 2009, 69:4941-4944). Because the presence of an Fc region and its effector functions shows improved in vivo properties for many therapeutic applications, a variety of Fc-containing bispecific antibody designs have also have been suggested (Coloma & Morrison, Nat Biotechnol 1997, 15:159-163; Shen et al., J Biol Chem 2006, 281:10706-10714; Asano et al., J Biol Chem 2007, 282:27659-27665; Wu et al., Nat Biotechnol 2007, 25:1290-1297).
We have developed a novel approach for constructing multivalent antibodies using the dock-and-lock (DNL) method (Rossi et al., Proc Natl Acad Sci USA 2006, 103:6841-6846), which enables site-specific self-assembly of two modular components only with each other. The DNL method results in a covalent structure of defined composition with retained bioactivity (Chang et al., Clin Cancer Res 2007, 13:5586s-5591s). Since the co-administration of anti-CD20 and anti-CD22 antibodies showed improved anti-lymphoma efficacy without increased toxicity in patients (Leonard et al., J Clin Oncol 2005, 23:5044-5051; Leonard & Goldenberg, Oncogene 2007, 26:3704-3713), and enhanced activity in a lymphoma xenograft model (Stein, et al., Clin Cancer Res 2004, 10:2868-2878), the studies reported in the present application utilized the DNL technique to develop hexavalent, monospecific anti-CD20 or bispecific anti-CD20/22 constructs with improved pharmacokinetic properties, increased efficacy and novel mechanisms of action for killing B-cell lymphoma, leukemia or other target cells.